Han Ridong, Peng Tao, Han Jiayu, Cui Hai, Liu Lu
College of Computer Science and Technology, Jilin University, Changchun, Jilin 130012, China; Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun, Jilin 130012, China.
College of Computer Science and Technology, Jilin University, Changchun, Jilin 130012, China; College of Software, Jilin University, Changchun, Jilin 130012, China; Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun, Jilin 130012, China.
Neural Netw. 2022 Aug;152:191-200. doi: 10.1016/j.neunet.2022.04.019. Epub 2022 Apr 21.
Wrong-labeling problem and long-tail relations severely affect the performance of distantly supervised relation extraction task. Many studies mitigate the effect of wrong-labeling through selective attention mechanism and handle long-tail relations by introducing relation hierarchies to share knowledge. However, almost all existing studies ignore the fact that, in a sentence, the appearance order of two entities contributes to the understanding of its semantics. Furthermore, they only utilize each relation level of relation hierarchies separately, but do not exploit the heuristic effect between relation levels, i.e., higher-level relations can give useful information to the lower ones. Based on the above, in this paper, we design a novel Recursive Hierarchy-Interactive Attention network (RHIA) to further handle long-tail relations, which models the heuristic effect between relation levels. From the top down, it passes relation-related information layer by layer, which is the most significant difference from existing models, and generates relation-augmented sentence representations for each relation level in a recursive structure. Besides, we introduce a newfangled training objective, called Entity-Order Perception (EOP), to make the sentence encoder retain more entity appearance information. Substantial experiments on the popular New York Times (NYT) dataset are conducted. Compared to prior baselines, our RHIA-EOP achieves state-of-the-art performance in terms of precision-recall (P-R) curves, AUC, Top-N precision and other evaluation metrics. Insightful analysis also demonstrates the necessity and effectiveness of each component of RHIA-EOP.
错误标注问题和长尾关系严重影响远距离监督关系抽取任务的性能。许多研究通过选择性注意力机制减轻错误标注的影响，并通过引入关系层次结构来共享知识以处理长尾关系。然而，几乎所有现有研究都忽略了这样一个事实：在一个句子中，两个实体的出现顺序有助于对其语义的理解。此外，它们仅分别利用关系层次结构的每个关系级别，而没有利用关系级别之间的启发式效应，即更高级别的关系可以为较低级别的关系提供有用信息。基于上述情况，在本文中，我们设计了一种新颖的递归层次交互注意力网络（RHIA）来进一步处理长尾关系，该网络对关系级别之间的启发式效应进行建模。从顶部到底部，它逐层传递与关系相关的信息，这是与现有模型最显著的区别，并以递归结构为每个关系级别生成关系增强的句子表示。此外，我们引入了一种全新的训练目标，称为实体顺序感知（EOP），以使句子编码器保留更多实体出现信息。我们在流行的《纽约时报》（NYT）数据集上进行了大量实验。与先前的基线相比，我们的RHIA-EOP在精确率-召回率（P-R）曲线、AUC、Top-N精确率和其他评估指标方面取得了领先的性能。有见地的分析还证明了RHIA-EOP每个组件的必要性和有效性。
